Pancreatic Cancer Detection in Whole Slide Images Using Noisy Label Annotations

Han, Ling; Samaras, Dimitris; Kurç, Tahsin; Gupta, Rajarsi; Kenneth, Shroyer; Saltz, Joel

doi:10.1007/978-3-030-32239-7_60

Cited by 38 publications

(17 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By contrast, much less attention has been given to noisy label learning in medical imaging 16,50 . Most existing studies concentrate on designing new loss weighting strategies 51,52 or new loss functions 53 . Since various issues with the datasets may exist in reality, the effectiveness of these methods is compromised 22,26 .…”

Section: Discussionmentioning

confidence: 99%

Annotation-efficient deep learning for automatic medical image segmentation

et al. 2021

View full text Add to dashboard Cite

Automatic medical image segmentation plays a critical role in scientific research and medical care. Existing high-performance deep learning methods typically rely on large training datasets with high-quality manual annotations, which are difficult to obtain in many clinical applications. Here, we introduce Annotation-effIcient Deep lEarning (AIDE), an open-source framework to handle imperfect training datasets. Methodological analyses and empirical evaluations are conducted, and we demonstrate that AIDE surpasses conventional fully-supervised models by presenting better performance on open datasets possessing scarce or noisy annotations. We further test AIDE in a real-life case study for breast tumor segmentation. Three datasets containing 11,852 breast images from three medical centers are employed, and AIDE, utilizing 10% training annotations, consistently produces segmentation maps comparable to those generated by fully-supervised counterparts or provided by independent radiologists. The 10-fold enhanced efficiency in utilizing expert labels has the potential to promote a wide range of biomedical applications.

show abstract

Section: Discussionmentioning

confidence: 99%

Annotation-efficient deep learning for automatic medical image segmentation

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Studies ( Cheerla and Gevaert, 2019 ; Peng et al., 2020 ) have demonstrated that WSI data alone, as well as together with genomic data, can achieve a remarkable performance in cancer prognosis prediction. However, most pancreatic cancer-specific studies using WSI data focused on diagnosis, i.e., pancreatic cancer detection and segmentation ( Fu et al., 2021 ; Kriegsmann et al., 2021 ; Le et al, 2019 ). The predictive value of WSI for prognosis purpose has not been rigorously shown in pancreatic cancer.…”

Section: Discussionmentioning

confidence: 99%

Robust deep learning model for prognostic stratification of pancreatic ductal adenocarcinoma patients

Wismans

Mustafa

et al. 2021

iScience

View full text Add to dashboard Cite

show abstract

“…The current mainstream approach is to train deep learning models directly on RGB images. But these works could learn invalid information from RGB images, and they require larger data sets or more complex neural networks to achieve better results [7]. To tackle this issue, we propose the nuclear density distribution feature (NDDF) to further improve the performance of the deep learning model.…”

Section: Methodsmentioning

confidence: 99%

Nuclear Density Distribution Feature for Improving Cervical Histopathological Images Recognition

Wang

Yang

Lyu

et al. 2021

2021 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Cervical carcinoma is a common type of cancer in the female reproductive system. Early detection and diagnosis can facilitate immediate treatment and prevent progression of the disease. However, in order to achieve better performance, DL-based algorithms just stack various layers with low interpretability. In this paper, a robust and reliable Nuclear Density Distribution Feature (NDDF) based on priors of the pathologists to promote the Cervical Histopathological Image Classification (CHIC) is proposed. Our proposed method combines the nucleus mask segmented by U-Net with the segmentation grid-lines generated from pathology images utilizing SLIC to obtain the NDDF map, which contains information about the morphology, size, number, and spatial distribution of nuclei. The result shows that the proposed model trained with NDDF maps has better performance and accuracy than that trained on RGB images (patch-level histopathological images). More significantly, the accuracy of the twostream network trained with RGB images and NDDF maps is steadily improved over the corresponding baselines of different complexity.

show abstract

Pancreatic Cancer Detection in Whole Slide Images Using Noisy Label Annotations

Cited by 38 publications

References 11 publications

Annotation-efficient deep learning for automatic medical image segmentation

Annotation-efficient deep learning for automatic medical image segmentation

Robust deep learning model for prognostic stratification of pancreatic ductal adenocarcinoma patients

Nuclear Density Distribution Feature for Improving Cervical Histopathological Images Recognition

Contact Info

Product

Resources

About